Compare AI models

Specifications
Parameters
8B/70B14.3B
Context window
8k32k
Benchmarks
HumanEval
81.7%34.2%
MMLU
82%62.5%
GSM8K
93%61.5%
Overview
CompanyMetaQwen
Release dateApr 18 2024Mar 28 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

LLaMA 3 (8B/70B) leads Qwen1.5-MoE on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). Qwen1.5-MoE shipped 21 days before LLaMA 3 (8B/70B), so benchmark comparisons should account for the intervening progress.

LLaMA 3 (8B/70B) has 8B/70B parameters, while Qwen1.5-MoE has 14.3B. Context windows are 8k (LLaMA 3 (8B/70B)) vs 32k (Qwen1.5-MoE).

On HumanEval, LLaMA 3 (8B/70B) leads at 81.7% vs Qwen1.5-MoE at 34.2%. On MMLU, LLaMA 3 (8B/70B) leads at 82% vs Qwen1.5-MoE at 62.5%. On GSM8K, LLaMA 3 (8B/70B) leads at 93% vs Qwen1.5-MoE at 61.5%.